HAL Id: hal-00839784
https://hal-polytechnique.archives-ouvertes.fr/hal-00839784
Submitted on 1 Jul 2013
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Reactivity differences between 2,4- and 2,5-disubstituted zirconacyclopentadienes: a highly selective and general
approach to 2,4-disubstituted phospholes.
Guillaume Bousrez, Florian Jaroschik, Agathe Martinez, Dominique Harakat, Emmanuel Nicolas, X. Le Goff, Jan Szymoniak
To cite this version:
Guillaume Bousrez, Florian Jaroschik, Agathe Martinez, Dominique Harakat, Emmanuel Nicolas, et al.. Reactivity differences between 2,4- and 2,5-disubstituted zirconacyclopentadienes: a highly selective and general approach to 2,4-disubstituted phospholes.. Dalton Transactions, Royal Society of Chemistry, 2013, epub ahead of print. �10.1039/c3dt51158h�. �hal-00839784�
Journal Name
Cite this: DOI: 10.1039/c0xx00000x www.rsc.org/xxxxxx
Dynamic Article Links►
ARTICLE TYPE
Reactivity differences between 2,4- and 2,5-disubstituted
zirconacyclopentadienes: a highly selective and general approach to 2,4- disubstituted phospholes.
Guillaume Bousrez,a Florian Jaroschik,*a Agathe Martinez,a Dominique Harakat,a Emmanuel Nicolas,b Xavier F. Le Goff,b Jan Szymoniak*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX DOI: 10.1039/b000000x
Mixtures of 2,4- and 2,5-disubstituted zirconacyclopentadienes were obtained by the reductive coupling of terminal alkynes using the Cp2ZrCl2/lanthanum system. Reactions of dihalophosphines with these mixtures afforded selectively the corresponding 2,4-disubstituted phospholes and 1,4-disubstituted
10
butadienes. The new series of phospholes was characterized by multi-nuclear NMR spectroscopy and X- ray analysis. A possible explanation for the observed selectivity was obtained from X-ray studies and DFT analysis of the intermediate zirconacyclopentadienes.
Introduction
Zirconacyclopentadienes are important intermediates in organic
15
synthesis, including transition-metal catalysis and macrocyclisation reactions. Since the first synthesis of tetraphenylzirconacyclopentadiene in 1961, several synthetic pathways have been explored. Especially, the emergence of low- valent zirconocene precursors, such as the Negishi, Takahashi or
20
Rosenthal reagents, has contributed considerably to their development. These complexes reductively couple internal alkynes with a large variety of substituents. The regioselectivity can be influenced by trimethylsilyl (alpha-position), pentafluorophenyl or mesityl groups (beta-position). The
25
hydrozirconation/methylation pathway developed by Buchwald is an interesting alternative as it gives readily access to trisubstituted complexes whereas the former reagents are especially suitable for tetrasubstituted compounds. However, most reagents cannot be employed in the coupling of two
30
terminal alkynes and disubstituted zirconacyclopentadienes have been rarely described. In most cases, mixtures of products were obtained which could not be further exploited. Only the very bulky tris(trimethylsilyl)silyl group afforded the 2,5-disubstituted zirconacyclopentadiene as sole product. A switch to the
35
bis(indenyl)zirconocene allows the 2,5-selective coupling of terminal alkynes. In 2006, the reductive dimerisation of terminal alkynes using a Mischmetall generated zirconocene(II) equivalent was described. These reactions yielded clean mixtures of 2,4- and 2,5-disubstituted zirconacyclopentadienes as shown by hydrolysis
40
experiments. To date no further reaction involving 2,4- disubstituted zirconacyclopentadienes has been reported.
Among the numerous transformations of zirconacyclopentadienes, the synthesis of 5-membered heterocycles has attracted much interest. The so-called Fagan-
45
Nugent route involves the reaction of electrophilic main group halides based on boron, phosphorus, tin and others with zirconacyclopentadienes. We focused our interest on the synthesis of new phospholes, as they have recently become important building blocks in the fields of catalysis, coordination
50
chemistry, material sciences and medicinal chemistry.1-5 In addition, phospholide anions are interesting alternatives to cyclopentadienide ligands in organometallic chemistry.6 In view of these numerous applications, it seemed surprising that not all substitution patterns of phospholes have been investigated: the
55
majority of phospholes is tetrasubstituted, with 2,5- and 3,4- disubstituted phospholes being also readily accessible.7 In contrast, very few examples of 2,4-disubstituted phospholes are known and no general synthetic route to such compounds has been reported.8 Such compounds may offer different steric and
60
electronic properties, which could be exploited in ligand design for catalytic and organometallic purposes.
We report here a highly selective and general synthesis of 2,4- disubstituted phospholes using the Fagan-Nugent route in combination with the reductive coupling of terminal alkynes by
65
the lanthanum generated zirconocene(II). An X-ray study and DFT calculations on the intermediate zirconacyclopentadienes offer some insights with respect to the observed selectivity.
Experimental part
All reactions were conducted under an argon atmosphere using
70
standard Schlenk techniques and an argon-filled Jacomex BS531- type dry box. Tetrahydrofuran and diethylether were collected under argon from a PURSOLV MD-3 (Innovative Technologie Inc.) solvent purification unit. Zirconocene dichloride was purchased from Strem Chemicals. Alkynes were purchased from
75
Aldrich and Alfa Aesar or synthetized from the corresponding aldehydes according to reported literature procedures.1
2 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]
Lanthanum ingot was purchased from Aldrich and Strem Chemicals and freshly filed in the drybox prior to use. 1H, 13C,
19F, 29Si and 31P NMR spectra were recorded in CDCl3, unless specified, on a 250MHz Bruker Avance I spectrometer equipped with a QNP probe and a 500MHz Bruker Avance III spectrometer
5
equipped with a BBFO+ probe. Chemical shifts are reported in delta (δ) units, expressed in parts per million (ppm). One and two dimensional experiments, including NOESY and HOESY (1H-
31P) experiments were performed for the NMR assignments of phospholes. High resolution ESI-MS spectra were recorded on a
10
hybrid tandem quadrupole/time-of-flight (Q-TOF) instrument, equipped with a pneumatically assisted electrospray (Z-spray) ion source (Micromass, Manchester, UK) operated in positive mode.
High resolution EI-MS spectra were obtained on a GCT-TOF mass spectrometer (Micromass, Manchester, UK) with EI source.
15
Synthesis of 1,4-diiodo-1,3- and 1,4-diphenylbutadienes 2a and 2b. A Schlenk tube was loaded with zirconocene dichloride (Cp2ZrCl2) (292 mg, 1.0 mmol), lanthanum (93 mg, 0.66 mmol) and THF (5 mL). The resulting mixture was stirred vigorously at room temperature until a deep red color appeared. At this stage,
20
phenylacetylene (0.21 ml, 2 mmol) was added to the reaction mixture. After 3 h, iodine (570 mg, 2.25 mmol) was added portion-wise at room temperature and the reaction was stirred for 24 h. The resulting brown solution with a yellow precipitate was quenched with aqueous Na2S2O3 solution (1 M, 5 mL) and water
25
(25 mL) at room temperature. The aqueous layer was extracted with Et2O (3 × 25 mL), and the organic phases were combined, washed with brine, dried over MgSO4, and concentrated under vacuum. The red residue was purified by flash column chromatography on silica gel using petroleum ether as eluent,
30
yielding 2a and 2b in 25% (113 mg, 0.25 mmol) and 18% (82 mg, 0.18 mmol), respectively. NMR data of 2a and 2b are in accordance with the literature data.5
Optimised procedure for 1,2,4-triphenylphosphole. A Schlenk tube was loaded with Cp2ZrCl2 (584 mg, 2.0 mmol), lanthanum
35
(186 mg, 1.3 mmol) and THF (10 mL). The resulting mixture was stirred vigorously at room temperature until a deep red color appeared. At this stage, the phenylacetylene (0.42 mL, 4.0 mmol) was added to the reaction mixture and the stirring was continued until complete disappearance of the alkyne as shown by TLC.
40
Then the optimized amount of dichlorophenylphosphine (0.14 mL, 1.0 mmol) was added at -78°C. After slow warming to room temperature, the reaction mixture was stirred for 18 h. After that time, petroleum ether (20 mL) was added to the brown solution and the solution was filtered over a short column of basic
45
aluminum oxide using petroleum ether/ethyl acetate 8:2 as eluent.
The solvent was evaporated and the crude residue was purified by flash column chromatography on silica gel using petroleum ether to yield phosphole 3a in 70% yield (218 mg, 70 mmol). Crystals suitable for X-ray analysis were obtained obtained by
50
recrystallisation from diethyl ether. 1H NMR (500 MHz, CDCl3):
7.08 (dd, JP-H = 40.0 Hz, JH-H = 1.5 Hz, 1H, H1), 7.22-7.29 (m, 3H, H12, H15), 7.31 (d, JH-H = 8.0 Hz, 2H, H11), 7.34-7.38 (m, 3H, H8, H16), 7.43 (d, JH-H = 7.0 Hz, 2H, H7), 7.46 (d, JP-H = 7.5 Hz, 2H, H14), 7.61 (d, JH-H = 8.0 Hz, 2H, H10), 7.66 (dd, JP-H =
55
12.5 Hz, JH-H = 1.5 Hz, 1H, H3), 7.73 (d, JH-H = 7.0 Hz, 2H, H6).
13C NMR (125 MHz, CDCl3): 126.6 (d, JP-C = 1.3 Hz, CH, C6), 126.8 (d, JP-C = 9.5 Hz, CH, C10), 127.5 (CH, C12), 128.0 (CH,
C1), 128.2 (CH, C8), 128.8 (CH, C7), 128.8 (d, JP-C = 6.6 Hz, CH, C15), 128.9 (CH, C11), 129.7 (d, JP-C = 1.5 Hz, CH, C16),
60
130.8 (d, JP-C = 9.9 Hz, C, C13), 131.9 (d, JP-C = 10.3 Hz, CH, C3), 134.0 (d, JP-C = 19.5 Hz, CH, C14), 136.5 (d, JP-C = 15.9 Hz, C, C9), 137.0 (d, JP-C = 3.0 Hz, C, C5), 150.4 (d, JP-C = 7.8 Hz, C, C2), 153.8 (d, JP-C = 2.0 Hz, C, C4). 31P NMR (200 MHz, CDCl3): 11.3. HRMS (EI) for C22H17P : calc. (m/z) 312.1068 ;
65
found (m/z) 312.1068.
Synthesis of 2,5-diphenylzirconacyclopentadiene 1b. A Schlenk tube was loaded with Cp2ZrCl2 (292 mg, 1.0 mmol), lanthanum (93 mg, 0.66 mmol) and THF (5 mL). The resulting mixture was stirred vigorously at room temperature until a deep
70
red color appeared. At this stage, phenylacetylene (0.21 mL, 2.0 mmol) was added to the reaction mixture and stirring continued for 3 h. The solution was filtered to remove insoluble LaCl3 salts.
Then the solvent was reduced to 2 mL under vacuum and put in the fridge at 4°C. From this solution a crop of dark red crystals of
75
1b suitable for X-ray diffraction was obtained. The 13C NMR data is in agreement with the analogous complex [(C5H4Me)2Zr(PhC(CH)2CPh)].4
1H NMR (250 MHz, C6D6): 5.91 (s, 10H), 7.03-7.06 (m, 5H), 7.13-7.14 (m, 2H), 7.27-7.33 (m, 5H). 13C NMR (62.5 MHz,
80
C6D6): 112.5 (CH), 119.6 (CH), 128.7 (CH), 129.8 (CH), 130.1 (CH), 130.5 (CH), 130.7 (CH), 149.9 (C), 192.7 (C). HRMS (ESI) for C26H23Zr [M+H]: calc. (m/z) 425.0847; found (m/z) 425.0853.
X-ray crystallography. Single crystals were coated in Paratone-
85
N oil and mounted on a loop. Data were collected at 150.0(1) K on a Nonius Kappa CCD diffractometer using a Mo Kα (λ=
0.71070 A) X-ray source and a graphite monochromator. All data were measured using phi and omega scans. The crystal structures were solved using SIR 97 and refined using Shelxl97.2,3
90
DFT-analysis. Calculations were performed with the Gaussian03 and Gaussian 09 suite of programs.6,7 The B3LYP functional8 was used for computation of monomers and NBO calculations as implemented in Gaussian 03, and the B97D functional9 for computation of the tetramer, as implemented in Gaussian09. The
95
lanld2z basis set and associated core potential10 were used for Zr atoms, and main group elements (C and H)11 were described with the 6-31G basis set. All basis sets were used as implemented in the program. Stationary points were identified as having no imaginary frequency. NBO analyses were performed using NBO
100
5.912 as implemented in Gaussian 03.
Reductive dimerization of terminal alkynes
The reductive dimerisation of phenylacetylene using the lanthanum generated zirconocene(II), yielded a dark red solution of a mixture of diphenylzirconacyclopentadienes 1a-c, as
105
evidenced by ESI-MS, X-ray crystallography (see below) and quenching experiments (Scheme 1). Hydrolysis of this mixture 1a-c showed that the ratio of the 1,3- and 1,4-disubstituted butadiene isomers was 1:1, whereas the 2,3-isomer was present in lower than 5% yield according to NMR analysis.‡ Trace amounts
110
of benzene derivatives were also present.11 The observed ratio for the butadienes is close to the one reported by Erker et al. obtained with a zirconocene-diene complex and phenylacetylene.
However, in our case no side-products arising from the zirconocene precursor were observed. This result was further
115
confirmed by the reaction of 1a-c with iodine. The 1,4- diiodobutadienes 2a and 2b were obtained in a 1:1 ratio according to the crude NMR spectrum. Full conversion was observed without the addition of copper catalyst.12 Compounds 2a and 2b were isolated in low 25% and 18% yield, respectively,
5
due to partial decomposition on silica gel. 2a was obtained previously only in trace amounts by the titanium-catalysed coupling of phenylacetylene.13
Ph
H
Cp2ZrCl2/La THF, t.a ., 4h
Cp2Zr
H+or I2
1 a: 2,4-Ph 1b: 2,5-P h 1 c: 3,4-Ph 2
Ph Ph
X X
X X
X X Ph
Ph Ph
Ph
Ph
Ph
X = H 1 1 0.05
X = I 1 (2a) 1 (2b) n.o.
+ +
Scheme 1 Hydrolysis and iodonolysis of disubstituted
10
zirconacyclopentadienes
Selective synthesis of 2,4-disubstituted phospholes
Further transformation of diiodide 2a via iodine/lithium exchange followed by addition of dichlorophenylphosphine yielded the first
15
2,4-diphenylsubstituted phosphole 3a in 65% yield (eq. 1).7a
1)n-BuLi
Et2O , -78°C to r.t., 1h 2) PhPCl2
-78°C to r .t., 1h
P Ph
3a P h Ph 2a
(1)
In order to simplify the procedure and circumvent the diiodide compounds, the addition of one equivalent of PhPCl2 to diphenylzirconacyclopentadienes 1a-c was investigated.
20
Surprisingly, this did not lead to a mixture of the expected 1,2,4- and 1,2,5-triphenylphospholes, but after workup 1,2,4- triphenylphosphole 3a and 1,4-diphenyl-1,3-butadiene 4a were obtained in a 1:1 ratio (eq. 2, table 1). The products could be separated by column chromatography and identified by
25
multinuclear NMR spectroscopy. Further improvement was achieved when only 0.55 equivalents of dichlorophenylphosphine was added as the formation of by-products due to unreacted phosphorus reagent could be avoided. Phosphole 3a was isolated as a pale-yellow, relatively air-stable solid in 70% yield based on
30
phosphorus reagent. This methodology was then extended to a large variety of phospholes carrying aryl 3b-f, heteroaryl 3g, alkyl 3h-i and trimethylsilyl 3j groups. An initial reaction was carried out with 1.0 equivalent of PhPCl2 to determine the ratio between phosphole 3 and butadiene 4. Based on this result, in a
35
second reaction the optimised amount of dichlorophosphine was added to prepare the phospholes 3. The isolated yields vary between 72 and 24% (Table 1) and in all cases, only the 2,4–
disubstituted phospholes were obtained. The substituents R on the
terminal alkyne have a limited influence on these ratios. The only
40
exception is the p-chlorophenyl substituent which clearly favors the formation of 2,4-disubstituted phosphole 3c.14 In order to further investigate this behaviour, the reaction with other electrophilic phosphorus reagents was examined. PhPBr2 has been described as being more reactive than PhPCl2.9a However,
45
when this reagent was used in the reaction with 1a-c, only compound 3a was formed. Heating at 60°C for 15h did not influence the reaction outcome. Reaction of 1a-c with dichlorocyclohexylphosphine afforded only the 2,4- diphenylphosphole 3k.
50
+ 1) Cp2ZrCl2/L a
THF, t.a., 4h 2) R'PCl2
-78°C to r .t., 2 4h P
R
R 3a-k R'
4 a-j
R H
2
H H
R
R
(2)
Table 1 Synthesis and 31P NMR data of 2,4-disubstituted phospholes Entry R R’ Ratioa 3/4 Yieldb 3 31P NMRc
a Ph Ph 1/1 70 (35) 11.3
b p-CH3-(C6H5) Ph 1/1 61 (30) 10.8 c
d e f g h i j k
p-Cl-(C6H5) p-F-(C6H5) p-OMe-(C6H5)
2-naphthyl 2-thienyl
n-pentyl t-butyl Me3Si Ph
Ph Ph Ph Ph Ph Ph Ph Ph c-C6H11
4/1 2/1 2/1 2/1 1/1 1/1 2/1 1/1 1/1
46 (39) 48 (34) 36 (26) 24 (17) 57 (29) 63 (32) 47 (33) 72 (36) 40 (20)
12.3 11.9 10.6 11.7 13.4 8.1 1.4 31.9 26.0
a determined by 1H NMR from the crude reaction mixture; b isolated yield under optimised conditions with respect to phosphorous reagent and in parentheses with respect to zirconium reagent; c in CDCl3.
55
In order to ensure that the observed reactivity did not originate from the employed zirconocene source Cp2ZrCl2/La, an experiment was conducted using phenylpropyne. As expected, in this case both phospholes 3l and 3m were formed in a 3:1 ratio in an overall 66% yield (eq. 3).
60
+ 1) Cp2Zr Cl2/La
THF, t.a., 4 h 2) PhP Cl2
- 78°C to r.t., 24h P Ph
Ph 3l Ph
Ph
Ph 3m P h
Me 2
Me
Me
Me Me P Ph
66 %,3l/3m= 3 /1
(3)
Finally, a competing experiment between di- and tetrasubstituted zirconacyclopentadienes coming from phenylacetylene 1a-c and from phenylpropyne 1d-e was performed. Addition of 0.5 equivalents of PhPCl2 to this mixture
65
4 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]
led to the exclusive formation of phosphole 3a. Further addition of 0.5 equivalents of PhPCl2 afforded a mixture of 3l and 3m, which further increased with more phosphorous reagent. The 2,5- disubstituted phosphole was not observed.
Cp2Zr
Cp2Zr Ph
Me
Ph Me Cp2Zr
P h Me
Me P h Cp2Zr
Ph
Ph Ph
Ph 0.5 0.5
0 .75 0.25
x P hPCl2
-78°C to r .t., 15h THF
3a + 3l + 3m
0 .5 : 0 : 0 x = 0.5 0.5 : 0 .4 : 0.1 x = 1.0 0 .5 : 0 .75 : 0 .25 x = 2.0
1 b 1a
1e 1d
+
5
Scheme 2 Competition experiment between di- and tetrasubstituted zirconacyclopentadienes
Characterisation of 2,4-disubstituted phospholes
Multinuclear NMR spectroscopy
The 31P NMR shifts of phospholes 3a-g are only slightly
10
influenced by the aryl substituents going from 10.6 ppm for the methoxy group to 13.4 ppm for the thienyl group. Compared to the corresponding 1,2,5-triphenylphosphole 5a,15 compound 3a displays a downfield shift of 10 ppm. In the 1H NMR spectra the main feature of phospholes 3 is the large 2JP-H coupling (40 Hz)
15
between phosphorus and the hydrogen on the alpha ring carbon atom. In the 13C NMR spectra, the 1JC-P coupling depends on the substitution, ranging from 0 to 4 Hz for hydrogen and aromatic substituents to 26 Hz for the trimethylsilyl group. Further differences in the NMR shifts between differently substituted
20
phospholes are summarized in Table 2.
X-ray analysis
Single crystals of 3a suitable for X-ray analysis were obtained by crystallisation from diethyl ether at 4°C (Fig. 1). Compound 3a displays many similar features compared to the 2,5-diphenyl
25
analogue 5a (table 3).16,17 The pyramidal phosphorus atom lies 0.065(1) Å below a least-square fitted phosphole plane, compared to 0.048(1) Å for 5a. The P-C bonds are in the range of single bonds, with a slight difference between the unsubstituted (1.787(2) Å) and the phenyl-substituted (1.815(2) Å) ring
30
carbons. The double bonds in the phosphole ring (1.360(2) Å and 1.355(2) Å) are localised. The phenyl rings are slightly twisted with respect to the phosphole plane (xx° and xx°) compared to 4.63(8)° and 7.99(7)° for 5a.
35
C
C' C
C'
R H
H R
P Ph
Table 2 Selected NMR data of some 2,4-disubstituted phospholes
R 1H (Hα, Hβ)a,b 13C (Cα, Cα’, Cβ, Cβ’)a,c Ph
(3a)
7.08 (40.0), 7.66 (12.5)
6.91 (38.0), 7.56 (12.5)
6.88 (38.5), 7.43 (11.5)
6.24 (41.0), 6.75 (15.0)
7.43 (40.5), 7.29 (18.5)
128.0 (0), 153.8 (2.0), 131.9 (10.3), 150.4 (7.8) p-OMe-(C6H5)
(3e) 2-thienyl
(3g) t-butyl
(3i) Me3Si
(3j)
124.5 (0), 153.3 (1.8), 130.2 (10.0), 150.0 (7.6)
124.9 (0), 147.0 (4.4), 130.5 (9.1), 143.8 (7.8) 122.7 (1.3), 161.8 (7.1), 130.9 (10.8), 165.8 (9.1) 149.4 (10.8), 150.3 (26.1),
147.4 (10.1), 154.8 (5.5)
a in CDCl3; b in parentheses 2JPH and 3JPH coupling constants; c in parentheses 1JPC and 2JPC coupling constants
40
Fig. 1 Molecular structures of 3a and 5a (50% probability ellipsoids).
Hydrogen atoms are omitted for clarity.
Table 3 Selected bond distances (Å) and angles (°) of 3a and 5a
3a 5a
P1-C1 P1-C4 P1-C17
1.787(2) 1.815(2) 1.828(2)
1.823(2) 1.824(2) 1.836(2) C1-C2
C2-C3 C3-C4 C1-P1-C4
1.360(2) 1.466(2) 1.355(2) 90.72(7) 105.91(7) 104.16(6) 0.15(25)
1.357(2) 1.439(2) 1.359(2) 91.34(8) C1-P1-C17
C4-P1-C17 C1-C2-C3-C4
104.33(7) 105.49(7) 0.37(25)
Disubstituted vs. tetrasubstituted zirconacyclopentadienes
5
It was previously observed that bulky alkyl ligands in the alpha position of zirconacyclopentadienes can hinder the formation of phospholes.18 However, the inertness of 2,5-diarylzircona- cyclopentadienes towards PhPCl2 was very intriguing as tetrasubstituted zirconacyclopentadienes carrying aryl groups in
10
the 2 and 5 positions react readily with PhPX2 as previously reported and shown above.9a,c In order to gain more insight into this selective transformation, we focused next on the intermediate zirconacyclopentadienes.
X-ray study
15
X-ray quality crystals of 1b were obtained from a THF solution of a mixture of 1a-c at 4°C (Fig. 2). Compound 1b contains two zirconacyclopentadiene complexes and one THF molecule in the unit cell. Bond lengths and angles involving zirconium are similar to other zirconacyclopentadienes, as for example in
20
Cp2Zr(C4H2(TTMS)2) and Cp2Zr(C4Ph4).9a,14b The most important feature of 1b is the nearly planar phenyl-zirconacyclopentadiene- phenyl system with twist angles ranging from 5.33° to 16.91°.
This is in stark contrast to tetrasubstituted zirconacyclopentadienes where the phenyl rings in the 2 and 5
25
positions are bent at 40-60° due to steric crowding, for example in Cp2Zr(C4Ph4) the twist angles are 48.69° and 54.34°.14b,c
Fig. 2. Molecular structure of 2,5-diphenylzirconacyclopentadiene 1b (50% probability ellipsoids), only one of the two complexes is shown. A
30
THF molecule and hydrogen atoms are omitted for clarity. Selected bond
distances (Å) and angles (°): Zr(1)-C(7) 2.244(2), Zr(1)-C(10) 2.248(2), Zr(1)-C(21) 2.465(3), Zr(1)-C(24) 2.523(2), C(7)-C(8) 1.359(3), C(8)-
C(9) 1.474(3), C(9)-C(10) 1.360(3), C(7)-Zr(1)-C(10) 86.74(7) When the isolated complex 1b was reacted with PhPCl2 in THF-
35
d8 no reaction was observed.
DFT-analysis
Based on the X-ray structure of 1b, DFT calculations using the Gaussian03 suite of programs were carried out. Initially, the B3LYP functional was used, in conjunction with the 6-31G basis
40
set for all main-group elements, and the lanl2dz basis set for Zr. It was shown that the structure of 1b could not be correctly calculated. A deviation in the dihedral angles of 10-20° was observed. Variation of basis sets and functionals did not improve the results. A closer look at the crystal structure revealed that
45
intermolecular interactions (π-stacking) lead to a tetrameric structure, i.e. four zirconacyclopentadienes form an independent tetramer. This structure could be correctly reproduced using the B97D functional, which takes into account dispersion effects for short distance interactions (Table 4).
50
Table 4 Comparison of X-ray data and DFT calculations on 1b
Dihedral anglea C5-C8/C35-C38 [°]
C7-C10/C33-C36 [°]
C9-C12/C31-C34 [°]
X-ray dataa 16.91/5.33 0.86/0.11 12.89/13.3
Monomerb 25.75
1.15 25.77
Tetramerc 13.33/5.56
1.34/0.21 13.33/7.91
aTwo distinct complexes in unit cell; b B3LYP//6-31G (H, C) and lanl2dz (Zr); c B97D//6-31++G (H), 6-31G (C) and lanl2dz (Zr)
In solution, it seems however more likely that the monomeric
55
structure is prevalent. The 2,4-diphenylzirconacyclopentadiene 1a was therefore also calculated with the B3LYP functional and an NBO analysis of 1a and 1b was performed. For comparison reasons the tetrasubstituted complexes 1d and 1e arising from phenylpropyne were also calculated (Table 5, Fig. 3 and 4).
60
Zr Cp2
1a,d 1b,e
R
R
3 2
1
Zr Cp2
R R
3 2
1
Table 5 Comparison of calculateda dihedral angles (°) of 1a,b,d,e
2,4-Ph,3,5-H (1a) 2,5-Ph,3,4-H (1b) 2,4-Ph,3,5-Me (1d) 2,5-Ph,3,4-Me (1e)
Angle 1 +35.44 +25.75 +89.95 +54.91
Angle 2 +2.90 +1.15 –1.39 +1.90
Angle 3 +31.43 +25.76 –56.62 +54.91
a B3LYP//6-31G (H, C) and lanl2dz (Zr)
6 | Journal Name, [year], [vol], 00–00 This journal is © The Royal Society of Chemistry [year]
X Fig. 3 HOMOs and optimised geometries of zirconacyclopentadienes
5
1a,1b, 1d and 1e
Zr Cp2
P h
Ph
1a
Zr Cp2
Ph 1b
P h -0.40 - 0.63 -0.4 1 -0.4 1
HOMO -5.026 eV -4.901 eV
Zr Cp2
Ph
P h
1d
Zr Cp2
Ph 1e
Ph -0.4 4 -0.38 - 0.42 - 0.42
HOMO -4.923 eV -4.988 eV
Fig.4 NBO analysis and HOMO energies of 1a, 1b, 1d and 1e
10
These calculations lead to the following observations: (i) the HOMOs of all complexes are on the zirconacyclopentadiene backbones, however the energy differences are too small to explain the observed selectivity; (ii) an important difference in the twist angles of the phenyl groups is observed between 1b
15
(25.75°) and 1e (54.91°); (iii) the unsubstituted α-carbon in 1a has a significantly more negative charge (-0.63) than the corresponding phenyl-substituted carbons in 1b (-0.41) and 1e (- 0.42) and the methyl-substituted carbon in 1d (-0.38). This difference in charge is in agreement with a more reactive
20
intermediate 1a compared to 1b,d,e. However, the complete inertness of 1b towards dihalophosphines seems to be mainly related to steric hindrance arising from the small tilt angle of the phenyl groups.
Conclusions
25
In conclusion a straightforward one-pot synthesis of a large series of 2,4-disubstituted phospholes has been described making use of the difference in reactivity between 2,4- and 2,5-disubstituted zirconacyclopentadienes towards electrophilic phosphorous reagents. An X-ray study and DFT analyses of the intermediate
30
zirconacyclopentadienes point towards steric reasons and only slight electronic influences to explain this selectivity. Further studies to extend this methodology to other 5-membered heteroles as well as the investigation of the chemistry of the new phosphole building blocks are currently under way.
35
We thank the CNRS, the Université de Reims and the Region Champagne-Ardenne for financial support and the platform PlAneT for technical support. Assistance from Miss Carine Machado (mass spectrometry) is acknowledged.
40
Notes and references
a Institut de Chimie Moléculaire de Reims, UMR CNRS 7312, Université de Reims, BP 1039, 51687 Reims, France. Fax: +33 326913166; Tel:
+33 326913244; E-mail [email protected] and [email protected]
45
b Laboratoire “Hétéroéléments et Coordinationˮ, Ecole Polytechnique and CNRS, 91128 Palaiseau Cedex, France
† Electronic Supplementary Information (ESI) available: Experimental and DFT data. CCDC [] See DOI: 10.1039/b000000x/
§ The dehydrohalogenation of 1-halogenophospholium salts worked in the case of 1,3-dimethylbutadiene (ref 8a), however, 1,3-diarylbutadienes are difficult to access and readily undergo homo-Diels Alder reactions.
‡ The ratio 2a/2b = 1/2.5 reported in 2006 (ref. 10) was based on GC analysis and did not take into account the facile homo-Diels Alder
5
reaction of the butadienes.
1 F. Mathey, Chem. Rev., 1988, 88, 429.
2 D. Carmichael, in Phosphorus (III) Ligands in Homogeneous Catalysis: Design and Synthesis, Chapt. 7: Phospholes, (Ed. P Kamer and P. W. N. M. van Leeuwen), Wiley VCH, 2012.
10
3 (a) Y. Matano and T. Imahori, Acc. Chem. Res., 2009, 42, 1193; (b) P. Le Floch, Coord. Chem. Rev., 2006, 250, 627.
4 (a) Y. Ren and T. Baumgartner, Dalton Trans., 2012, 41, 7792; (b) Y. Matano and T. Imahori, Org. Biomol. Chem., 2009, 7, 1258; (c) T.
Baumgartner and R. Réau, Chem. Rev., 2006, 106, 4681.
15
5 S. Urig, K. Fritz-Wolf, R. Réau, C. Herold-Mende, K. Toth, E.
Davioud-Charvet and K. Becker, Angew. Chem. Int. Ed., 2006, 45, 1881.
6 (a) S. Labouille, F. Nief, X.-F. Le Goff, L. Maron, D. R. Kindra, H.
L. Houghton, J. W. Ziller and W. J. Evans, Organometallics, 2012,
20
31, 5196; (b) F. Jaroschik, F.Nief, X.-F. Le Goff and L. Ricard, Organometallics, 2007, 26, 3552.
7 (a) F. Nief, B. Tayart de Borms, L. Ricard and D. Carmichael, Eur. J.
Inorg. Chem., 2005, 637; (b) J. Hydrio, M. Gouygou, F. Dallemer, G.
G. A. Balavoine and J.-C. Daran, Eur. J. Org. Chem., 2002, 4, 675;
25
(c) M. Ogasawara, K. Yoshida and T. Hayashi, Organometallics, 2001, 20, 1014;
8 (a) W. P. Freeman, Y. J. Ahn, T. K. Hollis, J. A. Whitaker, V.C.
Vargas, R. J. Rubio, K. D. Alingog, E. B. Bauer and F. S. Tham, J.
Organomet. Chem., 2008, 693, 2415; (b) G. Keglevich, T.
30
Chuluunbaatar, A. Dobó and L. Tőke, J. Chem. Soc., Perkin Trans. 1, 2000, 1495.
9 (a) O. Fadhel, Z. Benkö, M. Gras, V. Deborde, D. Joly, C. Lescop, L.
Nyulászi, M. Hissler and R. Réau, Chem. Eur. J., 2010, 16, 11340;
(b) Y. Matano, M. Nakashima and H. Imahori, Angew. Chem. Int.
35
Ed., 2009, 48, 4002; (c) C. Hay, M. Hissler, C. Fischmeister, J. Rault- Berthelot, L. Toupet, L. Nyulászi and R. Réau, Chem. Eur. J, 2001, 7, 4222. (d) P. J. Fagan, W. A. Nugent and J. C. Calabrese, J. Am.
Chem. Soc. 1994, 116, 1880.
10 C. Denhez, S. Médégan, F. Hélion, J.-L. Namy, J.-L. Vasse and J.
40
Szymoniak, Org. Lett., 2006, 8, 2945.
11 A. Joosten, M. Soueidan, C. Denhez, D. Harakat, F. Hélion, J.-L.
Namy, J.-L. Vasse and J. Szymoniak, Organometallics, 2008, 27, 4152.
12 C. Xi, S. Huo, T. H. Afifi, R. Hara and T. Takahashi, Tetrahedron
45
Lett., 1997, 38, 4099.
13 S. Yamaguchi, R.-Z. Jin, K. Tamao and F. Sato, J. Org. Chem., 1998, 63, 10060.
14 (a) V. H. Gessner, J. F. Tannaci, A. D. Miller and T. Don Tilley, Acc.
Chem. Res., 2011, 44, 435. (b) A. D. Miller, J. F. Tannaci, S. A.
50
Johnson, H. Lee, J. L. McBee and T. Don Tilley, J. Am. Chem. Soc., 2009, 131, 4917. (c) S. A. Johnson, F.-Q. Liu, M. C. Suh, S. Zürcher, M. Haufe, S. S. H. Mao and T. Don Tilley, J. Am. Chem. Soc., 2003, 125, 4199.
15 B. Lukas, R. M. G. Roberts, J. Silver and A. S. Wells, J. Organomet.
55
Chem., 1983, 256, 103.
16 W. P. Ozbirn, R. A. Jacobsen and J. C. Clardy, J. Chem. Soc. D, 1971, 1062.
17 A low-temperature (150 K) crystal structure of 5a was recorded.
18 F. Nief, D. Turcitu and L. Ricard, Chem. Commun., 2002, 1646.
60
